Pyridobenzothiazole derivatives as new chemotype targeting the HCV NS5B polymerase

Article abstract of DOI:10.1016/j.bmc.2011.11.061

Hepatitis C virus (HCV) infection has been recognized as the major cause of liver failure that can lead to hepatocellular carcinoma. Among all the HCV proteins, NS5B polymerase represents a leading target for drug discovery strategies. Herein, we describe

Full text of DOI:10.1016/j.bmc.2011.11.061

Bioorganic & Medicinal Chemistry 20 (2012) 866–876
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Pyridobenzothiazole derivatives as new chemotype targeting the HCV
NS5B polymerase
Giuseppe Manfroni a, , Francesco Meschini a, Maria Letizia Barreca a, , Pieter Leyssen b, Alberta Samuele c,
⇑
⇑
Nunzio Iraci a, Stefano Sabatini a, Serena Massari a, Giovanni Maga c, Johan Neyts b, Violetta Cecchetti a
a Dipartimento di Chimica e Tecnologia del Farmaco, University of Perugia, 06123 Perugia, Italy
b Rega Institute for Medical Research, Katholieke Universiteit Leuven, B-3000 Leuven, Belgium
c Istituto di Genetica Molecolare, IGM-CNR, 27100 Pavia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 June 2011
Revised 25 November 2011
Accepted 26 November 2011
Available online 3 December 2011
Hepatitis C virus (HCV) infection has been recognized as the major cause of liver failure that can lead to
hepatocellular carcinoma. Among all the HCV proteins, NS5B polymerase represents a leading target for
drug discovery strategies. Herein, we describe our initial research efforts towards the identification of
new chemotypes as allosteric NS5B inhibitors. In particular, the design, synthesis, in vitro anti-NS5B
and in cellulo anti-HCV evaluation of a series of 1-oxo-1H-pyrido[2,1-b][1,3]benzothiazole-4-carboxylate
derivatives are reported. Some of the newly synthesized compounds showed an IC50 ranging from 11 to
Keywords:
Hepatitis C
NS5B inhibitors
HCV inhibitors
Pyridobenzothiazole derivatives
Thumb site I
23
lM, and molecular modeling and biochemical studies suggested that the thumb domain could be the
target site for this new class of NS5B inhibitors.
Ó 2011 Elsevier Ltd. All rights reserved.
Thumb site II
Molecular docking
1. Introduction
and promising direct-acting anti-virals (DAAs)—also named ‘spe-
cifically targeted anti-viral therapy for hepatitis C’ (STAT-C) com-
pounds—have been progressed in phase I to II clinical trials,6–8
with the NS3/4A protease inhibitors boceprevir (Victrelis™ from
Merck)9 and telaprevir (Incivek™ from Vertex Pharmaceutical),10
that very recently have reached the market. However, the use of
these protease inhibitors still requires the combination with the
Chronic hepatitis C virus (HCV) infection is emerging as a major
health burden with an estimated 180 million chronically infected
individuals worldwide and 3–4 million new cases of infection each
year.1,2 HCV has been found to be a major cause of cirrhosis, hepa-
tocellular carcinoma, as well as liver failure, and is the most com-
mon cause of the need for liver transplantation.3 To date, there is
no vaccine available against HCV. The standard treatment is a com-
aspecific pegylated interferon-a and ribavirin drugs. Beside NS3/
4A, one of the most interesting targets for the development of
DAAs is the viral protein NS5B, a key enzyme in the virus replica-
tion cycle having RNA-dependent RNA polymerase activity, a func-
tionality which is not present in human cells thus making the
target attractive. The discovery of a drug that targets NS5B poly-
merase still remains an exciting challenge that would help in the
development of an interferon/ribavirin-free HCV therapy, plausibly
based on a combination of DAAs acting on different targets to ham-
per rapid emergence of resistance. The NS5B polymerase offers a
wide range of possibilities for the discovery of new molecular enti-
ties, since it contains one active site and at least five allosteric dru-
gable sites.11
bination of pegylated interferon-a and ribavirin, a broad spectrum
anti-viral agent;4 this therapy is however aspecific, expensive,
poorly tolerated and of limited efficacy in genotype 1 infected pa-
tients, who comprise the majority of individuals that are chroni-
cally infected in Europe and North America. The development of
new, virus-specific, more efficacious, and better tolerated anti-viral
agents is therefore of primary importance.5
In the last decade, the acquired knowledge of the HCV life cycle
and of structural features of the HCV proteins have revealed sev-
eral targets for the development of potential novel therapeutics
Two main types of NS5B inhibitors have been developed to
date:12–14 nucleoside or nucleotide inhibitors (NIs) which bind
the active site of the enzyme, and non-nucleoside inhibitors (NNIs)
which target one of the allosteric sites of the enzyme preventing a
conformational transition needed for initiation of RNA synthesis.
⇑
Corresponding authors. Tel.: +39 075 585 5126; fax: +39 075 585 5115 (G.M);
tel.: +39 075 585 5157; fax: +39 075 585 5115 (M.L.B.).
E-mail addresses: giuseppe.manfroni@unipg.it (G. Manfroni), lbarreca@unipg.it
(M.L. Barreca).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.11.061

G. Manfroni et al. / Bioorg. Med. Chem. 20 (2012) 866–876
867
As a part of our effort to identify new anti-HCV chemotypes, we
have herein focused our attention on the inhibitors binding to the
allosteric site referred to as thumb site I (TSI), which is located at
the upper section of the thumb domain, approximately 30 Å from
the active site. Benzimidazole and indole derivatives represent
promising TSI non-nucleoside inhibitors (TSI-NNIs), and they have
been the focus of intense lead optimization research at several
pharmaceutical companies.12 As an example, exhaustive structural
modifications around the benzimidazole scaffold, performed by
Japan Tobacco, led to the identification of candidate JTK-109
(1)15(Fig. 1), a potent and selective inhibitor which was advanced
into phase II clinical trials and later terminated for undisclosed
reasons.
synthesized compounds indicate that this chemical class might
serve as a good starting point for further lead optimization
strategies.
2. Results and discussion
2.1. Rational design
The aim of our work was the identification of new chemotypes
as potential TSI-NNIs through the application of a traditional scaf-
fold hopping strategy. Scaffold hopping is a powerful tool to dis-
cover structurally novel compounds by modifying the central
core structure of known active compounds.16 Structural and struc-
ture–activity relationship (SAR) studies of known TSI-NNIs illus-
trated in the literature15,17,18 have pointed out the presence of
three key chemical features correlated with the NS5B inhibition
Herein, a series of pyridobenzothiazole derivatives were de-
signed and synthesized following a scaffold hopping strategy. In vi-
tro anti-NS5B and in cellulo anti-HCV evaluation of the newly
Figure 1. Chemical structure of JTK-109 (1) and design concept. The key chemical features are highlighted in red.
Scheme 1. Reagents and conditions: (i) cyclohexanol or isopentanol, PPh3, DEAD, TEA, THF, ultrasounds, rt; (ii) 50% NaOH, ethylene glycol, reflux; (iii) ethyl cyanoacetate,
120 °C; (iv) Vilsmeier reagent 90 °C; (v) appropriate anhydrides or diketene 100–110 °C; (vi) 4% NaOH, MeOH, 75 °C; (vii) aqueous 1 N LiOH, MeOH, 40 °C; (viii) m-CPBA,
CH2Cl2, rt.

868
G. Manfroni et al. / Bioorg. Med. Chem. 20 (2012) 866–876
ability, i.e., two lipophilic portions (HY1 and HY2) and one carbox-
ylic/carbonyl function (CF) able to establish salt bridge/hydrogen
bond interaction. These functionalities are maintained in a spatial
arrangement suitable for the interaction within the NS5B TSI by
a heterocyclic central core, typically an indole or a benzimidazole
ring. Additional branching moieties can be present to further opti-
mize potency and/or physiochemical properties. All these chemical
regions are illustrated in Figure 1, taking as example the benzimid-
azole derivative 1.
Our scaffold hopping drug design of new TSI-NNI chemotypes
was primarily focused on the identification of a heterocycle core
framework able to replace the central chemical template of
bioactive compounds such as 1, while keeping the above men-
tioned key interaction features in a spatial orientation appropriate
to fit into TSI.
To achieve this goal, a variety of scaffolds was scrutinized by a
thorough bibliographic research, looking for a heterocyclic system
never employed in the development of anti-HCV compounds, easy
to synthesize and suitable for appropriate chemical functionaliza-
tion. The ethyl 1-oxo-1H-pyrido[2,1-b][1,3]benzothiazole-4-car-
boxylate (2), reported by Kato et al.19 seemed to match all the
above mentioned requirements (Fig. 1). Indeed, the compound
presented an ester group easily hydrolyzable to the necessary CF,
and an HY1 portion embedded in the tricyclic system. Moreover,
the second required lipophilic area (HY2) could have been easily
inserted at the C-2 position adapting the procedure reported by
Kato (Fig. 1).
Thus, the pyridobenzothiazole derivatives 2a–h, characterized
by a different substitution pattern at C-2 and C-8 positions
(Scheme 1 and Table 1), were synthesized. Cyclohexyl (2a, 2b) or
phenyl groups (2c–f) were considered as lipophilic substituents
at the C-2 position; C-2 acetyl (2g) and C-2 unsubstituted (2h)
derivatives were also synthesized to further explore the
importance of the lipophilic area HY2.
Likewise
1 and its analogues, most of the synthesized
pyridobenzothiazoles have an additional aliphatic branching
moiety at the C-8 position such as a cyclohexyloxy (2b, 2c, 2g,
and 2h) or a isopentyloxy (2d) group; the presence of an aryl
moiety at the C-8 position, such as a phenyloxy, was also planned
but unfortunately attempts to synthesize the corresponding target
derivatives failed.20 To extend the SAR,
a smaller lipophilic
methoxy group (2e) or a fluorine atom (2f) were also considered
as C-8 substituents; the oxidation of sulphur atom to sulfone group
was also realized for some derivatives (3c, 3e and 3f).
Table 1
Biological evaluation of the pyridobenzothiazole derivatives realized in this study
O
X
On
S
R8
N
R2
O
R2
R8
H
X
n
NS5B IC50
(lM)
Replicon Huh 5-2
M) CC50 (lM)
a
Compds
b
b
EC50
(l
2a
2b
OH
OH
0
0
>40
53.5
>100
O
23 3c
>56.5
>100
55.8
>56.5
>100
>100
O
O
O
2c
3c
OH
OH
0
2
11 2c
19 3c
2d
2e
3e
2f
OH
OH
OH
OH
OH
0
0
2
0
2
>40
>40
>40
>40
>40
54.0
51.2
89.7
>100
66.0
55.0
51.8
>100
>100
>100
OMe
OMe
F
3f
F
O
2g
COMe
H
OH
OH
0
0
>40
>100
>100
O
2h
>40
>100
NDd
>100
ND
ATA
15
3
a
IC50 = concentration of compound that inhibits enzyme activity of 50%.
EC50 = the effective concentration required to inhibit virus induced cytopatic effect by 50% in a single experiment. CC50 = is the concentration required to reduce the
b
bioreduction of MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxy-phenyl)-2-(4-sulfophenyl)-2H-tetrazolium) in to formazan by 50%.
c
Mean values for at least three independent experiments standard deviation.
Not determined.
d

G. Manfroni et al. / Bioorg. Med. Chem. 20 (2012) 866–876
869
2.2. Chemistry
2.3.2. Anti-HCV activity
The designed compounds were also evaluated in Huh-5-2
cells, hepatocyte cell line stably carrying genotype 1b
subgenomic HCV replicon. The anti-viral activity (EC50) was
measured by the reduction of luciferase activity in the presence
of increasing amount of inhibitor, while the antimetabolic
The synthetic procedure for the preparation of the target
pyridobenzothiazole derivatives 2a–h, 3c, 3e, and 3f is depicted
in Scheme 1. Thus, commercially available 5-hydroxybenzothiaz-
ole was reacted with cyclohexanol or isopentanol in the usual
Mitsunobu reaction conditions and using ultrasound irradiation.
The obtained 5-functionalized benzothiazoles 4a and 4b, as well
as the commercially available 5-methoxy- and 5-fluoro-2-methyl-
benzothiazole were then demolished into the corresponding
aminothiophenol intermediates 5a, 5b, 5c,21 and 5d21 using a
mixture of aqueous 50% NaOH and ethylene glycol (1:1, v/v) at
reflux. Due to the tendency of thiol groups to oxidize by air
exposure to the corresponding non-reacting disulfur derivative,
the crude aminothiophenol intermediates were immediately
reacted with ethyl cyanoacetate to provide benzothiazole deriva-
tives 6a–d. In the same way, reaction of ethyl cyanoacetate with
commercially available aminothiphenol 5e provided benzothia-
zole derivative 6e.19
a
a
effects (CC50
(Table 1).
) was determined employing the MTS assay
Most pyridobenzothiazole derivatives proved to be not toxic
with the exception of the compounds 2b, 2d and 2e. A measur-
able anti-viral activity associated with no toxicity was obtained
for the sulfone derivative 3c (EC50 = 55.8
lM) and for sulphur
compound 2a (EC50 = 53.5 M). However, based on the hit selec-
l
tion criteria (>>70% inhibition has to be observed at concentra-
tions of compound that show no significant anti-metabolic
activity) none of them could be withheld as highly potent
selective inhibitor of HCV replication in the Huh-5-2 replicon
system.
The benzothiazole intermediates 6a–e were then reacted with
Vilsmeier reagent, prepared in situ by POCl3 and DMF, to obtain
the enamine intermediates 7a–e. These were subsequently con-
verted into pyridobenzothiazole esters 8a–h by reaction with
cyclohexylacetic anhydride (8a, 8b), phenylacetic anhydride (8c–
f), diketene (8g) or acetic anhydride (8h) in neat conditions. The
hydrolysis of the ester function, in a mixture of MeOH and aqueous
4% NaOH, gave the corresponding target acids 2a–f and 2h, while a
mild hydrolysis condition using 1 N LiOH and MeOH was required
to achieve the target acid 2g in order to avoid the formation of sev-
eral side-products. Starting from compounds 2c, 2e and 2f, the cor-
responding sulfone derivatives 3c, 3e and 3f were obtained by
oxidation with m-CPBA.
2.4. Binding mode analysis of hit compounds
The pyridobenzothiazole NS5B inhibitors 2b, 2c and 3c were
submitted to docking calculations in order to investigate their pos-
sible binding mode to the NS5B protein, and gain useful informa-
tion for a further chemical optimization process.
Since the compounds arose from a scaffold-hopping approach
starting from indole-based TSI-NNIs, TSI was first considered in
our docking study.
However, given the above mentioned mismatch of our preli-
minary SAR results with the SAR studies reported in literature for
different TSI-NNIs and the observation that the SAR of our com-
pounds got somehow overlapping to the thumb site II (TSII) NNIs
SAR, the possible binding modes of our pyridobenzothiazole deriv-
atives to TSII was investigated as well. It is worth noting that be-
side TSI-NNIs, TSII-NNIs require the presence of two hydrophobic
groups and a hydrogen bond acceptor as well.14
2.3. Biological evaluation
2.3.1. Enzymatic assay
The automated docking simulations were performed using the
software Glide version 5.0.24–27
The target compounds were tested for their ability to suppress
the in vitro incorporation of [3H]-UTP by recombinant HCV RNA
In the absence of disclosed structural information about the 1/
NS5B interaction, the crystal structure of the HCV polymerase com-
plexed with the indole-based TSI-NNI 9 (Fig. 2A) was used as target
protein(PDBID2BRK)28 toexplorethefirstthumbpocket. Thisstruc-
ture showed that the ligand binds to a site on the surface of the
thumb domain, mainly establishing hydrophobic interaction with
the surrounding NS5B residues. Indeed, the cyclohexyl group filled
in a deep pocket formed by aminoacid residues of Leu-392, Ala-
395, Thr-399, Ile-424, Leu-425, His-428 and Phe-429, while the phe-
nyl ring substituent filled in a narrow pocket surrounded by Val-37,
Leu-392, Ala-393, Ala-396, Leu-492 and Val-494. The indole core
bridged the two groups and also contributed both in filling the pock-
ets and in the interaction with the protein. Finally, the carboxylate
moiety formed a salt bridge with the exposed Arg-503 side chain.
First of all, test docking calculations using 9 were carried out to
validate the program performance. The ligand was thus extracted
from the corresponding NS5B complex and then docked back into
the allosteric pocket of the enzyme crystal structure. The best
docking pose of 9 (XP GScore = ꢀ9.58) agreed well with its exper-
imental binding conformation, with a root-mean square (rms)
deviation value of 0.3 Å.
polymerase NS5BDC21 (genotype 1b) on a homopolymeric RNA
primer/template (rA40/rU20) (Table 1). The non-nucleoside inhib-
itor aurintricarboxylic acid (ATA) was used as reference compound
in the enzymatic assay. This compound has been shown to inhibit
NS5B both in vitro and in replicon assays, through binding to the
benzothiadizine allosteric pocket.22
Compounds 2b, 2c, and 3c were able to significantly inhibit
the NS5B activity with IC50 values ranging from 11 to 23 lM.
These results highlighted that our rational design approach was
successful in finding new chemotypes as NS5B inhibitors. In
particular, the best anti-NS5B activity was observed for com-
pound 2c, functionalized with a phenyl at the C-2 and a cyclo-
hexyloxy group at C-8 positions. The data analysis suggested
that the simultaneous presence of two bulky hydrophobic groups
at position C-2 and C-8 was essential to ensure anti-NS5B activity
in the pyridobenzothiazole family. On the other hand, the oxida-
tion state of the sulphur atom seemed to have no influence on the
inhibitory ability, as showed by the comparable potency between
sulfone derivative 3c (IC50 = 19
derivative 2c (IC50 = 11 M).
3 lM) and its parent unoxidized
2
l
However, these preliminary SAR remarks do not completely
agree with SAR studies reported in literature for indole/benzimid-
azole TSI-NNIs, where a cyclohexyl group at the C-3 or N-1 position,
respectively, seems essential to have high anti-NS5B activity;23 in
fact, in the pyridobenzothiazole series the presence of a cyclohexyl
(2b) ora phenyl group (2c and 3c) at the corresponding C-8 position
resulted in compounds with a comparable potency.
Later, docking of compounds 2b, 2c and 3c in their anionic form
(MoKa29-predicted pKa of COOH was 4.08) generated similar bind-
ing modes, which closely resembled the crystallographic position
of the known inhibitor 9. As example, Figure 2B illustrates the
docked conformation of 3c (XP GScore = ꢀ7.33), the most interest-
ing hit of the pyridobenzothiazole series. In fact, the pyridobenzo-
thiazole carboxylate group (CF) interacted with the Arg-503 side

870
G. Manfroni et al. / Bioorg. Med. Chem. 20 (2012) 866–876
Figure 2. (A) Chemical structure of indole TSI-NNI 9. (B) Predicted binding mode of 3c (magenta) compared to the crystallographic position of 9 (green). Protein surface is
coloured according to the aminoacid side chains: basic, blue; acidic, red; polar, orange; nonpolar, gray.
chain, while the phenyl moiety (HY2) well overlapped with the
cyclohexyl ring of compound 9. Unfortunately, a visual analysis of
the docking poses also highlighted that our hit compounds were
not able to properly match the hydrophobic aromatic feature HY1
of known TSI-NNIs, since the benzene ring of the pyridobenzothiaz-
ole core was about 2 Å far away from the required functionality.
On the other hand, docking studies were also performed on 1
to explore the hypothesis that the extended hydrophobic portion
of our compounds might correspond to the extra hydrophobic
functionality of the known inhibitor. The modelling results
showed that 1 (XP GScore = ꢀ11.33) interacted with the NS5B res-
idues of the allosteric TSI pocket in a fashion similar to 9 (Fig. 3A),
with its branching moiety occupying an additional cleft mainly
defined by the residues Met-36, Val-37, Cys-146, Val-147 and
Leu-492.
In agreement with our design, the comparison between the
docking poses of 2a, 2b and 3c and the predicted binding mode
of 1 highlighted that the above described additional feature occu-
pied the same region within the TSI, being the cyclohexyl portion
of our hits well overlapped with one of the branching aromatic
Figure 3. (A) Docking conformation of 1 (cyan) compared to the experimental position of 9 (green). (B) Docking solutions of 1 (cyan) and 3c (magenta). Protein surface is
coloured according to the aminoacid side chains: basic, blue; acidic, red; polar, orange; nonpolar, gray.
Figure 4. (A) Chemical structure of filibuvir (10) (B) Predicted binding mode of 3c (magenta) compared to the crystallographic position of 10 (green). Protein surface is
coloured according to the aminoacid side chains: basic, blue; acidic, red; polar, orange; nonpolar, gray.

G. Manfroni et al. / Bioorg. Med. Chem. 20 (2012) 866–876
871
moiety of 1, as shown in Figure 3B for the pyridobenzothiazole
derivative 3c.
The crystal structure of NS5B polymerase in complex with PF-
00868554 (10, filibuvir, Pfizer, Fig. 4A) (PDB code 3FRZ),30 one of
the most advanced TSII-NNIs currently in clinical trials, was in-
stead used in the present study in order to explore the possible
interactions of 2b, 2c and 3c within the second thumb pocket.
Structural analysis showed that the binding of the hydroxydihy-
dropyranone derivative 10 to NS5B occurred in a predominantly
hydrophobic shallow pocket at the base of the thumb domain,
spacing about 35 Å from the active site. The cyclopentyl group
showed key contacts with residues Leu-419, Met-423, Tyr-477,
and Trp-528, whereas the two ethyl groups of the pyridine ring
interacted with Leu-419, Met-423, Ile-482, Val-485, Ala-486, Leu-
489, and Leu-497. A hydrogen bonding network was found be-
tween the dihydropyrone carbonyl and the donor–donor motif of
the protein (backbone NH of Ser-476 and Tyr-477) through one di-
rect hydrogen bond and one water-mediated hydrogen bond.
Moreover, the enol oxygen of the inhibitor formed a second
water-mediated hydrogen bond with Arg-501. Finally, the triazol-
opyrimidine, connected to the dihydropyrone core through a
Figure 5. Dependence of the inhibitory activity of compounds 2b, 2c and 3c from
the RNA concentration. Inhibition of NS5B incorporation of UTP on the (rA)40/
(rU)20, primer template was tested in the presence of a fixed subsaturating (i.e.,
<IC50) dose of inhibitor and two increasing doses of RNA template. Values are
expressed as % of inhibition with respect to control reactions in the absence of the
compounds. Plotted values are means of three independent determinations. Error
bars are S.D.
methylene linker, formed a p–p stacking interaction with His-475.
At first, we validated the ability of the docking algorithm to
reproduce the co-crystallized pose of 10 within this NS5B allosteric
site. This test yielded a good agreement between the in-silico
docked and the crystal structure, as pointed out by the obtained
rms deviation value of 0.68 Å. We next docked the three new
NS5B inhibitors into the 10 binding site of NS5B. The analysis of
the results highlighted that, analogously to what observed in the
docking studies on TSI, 2b, 2c and 3c generated reasonable binding
modes, similar to 10, within the second thumb pocket. For in-
stance, Figure 4B shows the obtained docking pose for compound
3c (XP GScore value = ꢀ7.43). The pyridobenzothiazole carboxylate
group (CF) formed the same hydrogen bonding network with Ser-
476 and Tyr-477 as observed for the dihydropyrone carbonyl group
of 10. The phenyl moiety (HY2) well overlapped with the cyclopen-
tyl ring of the co-crystallized inhibitor, whereas the aromatic fea-
ture HY1 and the extended hydrophobic portion of our
compounds showed the same hydrophobic interactions observed
for the two ethyl groups of 10.
our compounds interfere with the interaction between NS5B and
the RNA template.
3. Conclusions
The rational approach employed for the design of novel pyr-
idobenzothiazole-based TSI-NNIs led to the identification of new
hit compounds (2b, 2c and 3c) showing interesting inhibitory
activities in the enzymatic assay. Unfortunately, their cell-based
anti-HCV activities do not fully reflect the observed NS5B enzy-
matic inhibition; it may be supposed that the lack of remarkable
cellular activity could be due to permeability issue. Molecular
docking results suggested that the new NS5B inhibitors could bind
either TSI or TSII, despite the difference in scaffold and substitu-
tions. Biochemical studies provided further clues about a possible
mechanism of allosteric inhibition, showing that the pyridobenzo-
thiazole derivatives are non-competitive inhibitors with respect to
the ribonucleotide substrate and competitive inhibitors with re-
spect to the RNA template. Using the information mined from this
study and with the aim to improve both anti-NS5B and cell-based
anti-HCV activities, the design and synthesis of novel derivatives
will be the next step of this drug discovery project. Afterwards,
once stronger binders will be obtained, our efforts will be dedi-
cated to experimentally provide full proof of the exact binding site
for the pyridobenzothiazole derivatives.
Given that, as mentioned above, the two allosteric thumb sites
have similar requirements for inhibitor binding, it was not surpris-
ing to observe that our compounds could potentially exhibit simi-
lar affinity towards either TSI or TSII of NS5B. A thorough in-silico
analysis of the docking models of the pyridobenzothiazole deriva-
tives within both pockets could serve as a guide for future design
and development of more potent NS5B inhibitors.
2.5. Mechanism of inhibition study
4. Experimental section
Given the initial design principle and the molecular modeling
results, the mechanism of inhibition of our compounds could be
due to the perturbation of the dynamic properties of the thumb do-
main.31 For example, inhibitor binding could interfere with the
thumb b-loop motion, which has been proposed to interact with
the RNA template during elongation. In addition, residues in the
thumb-palm junction might contact the nascent primer strand. Fi-
nally, the lysine-arginine rich domain of the thumb has been pro-
posed to have a role in NS5B-RNA contacts. In order to define the
mechanism of inhibition, the potency of compounds 2b, 2c and
3c was evaluated in the in vitro assay, in the presence of varying
concentrations of either UTP or the RNA primer/template. No effect
of UTP concentration was observed on the inhibition (data not
shown). On the contrary, the potency of all compounds was de-
creased by increasing the RNA template/primer concentration
(Fig. 5). These results suggest that the allosteric effects caused by
4.1. Synthesis and compound characterization
All starting materials were commercially available, unless
otherwise indicated. Reagents and solvents were purchased from
common commercial suppliers and were used as such. Organic
solutions were dried over anhydrous Na2SO4 and concentrated
with a rotary evaporator at low pressure. All reactions were rou-
tinely checked by thin-layer chromatography (TLC) on silica gel
60F254 (Merck) and visualized by using UV or iodine. Column
chromatography separations were carried out on Merck silica gel
60 (mesh 70–230), flash chromatography on Merck silica gel 60
(mesh 230–400). Ultrasound reactions were conducted using
Decon ultrasonic LTD machinery. Melting points were determined
in capillary tubes (Büchi Electrotermal model 9100) and are uncor-
rected. Yields were of purified products and were not optimized.

872
G. Manfroni et al. / Bioorg. Med. Chem. 20 (2012) 866–876
1H NMR spectra were recorded at 200 or 400 MHz (Bruker Avance
DRX-200 or 400, respectively) while 13C NMR spectra were re-
corded at 100 MHz (Bruker Avance DRX-400). Chemical shifts are
given in ppm (d) relative to TMS. Spectra were acquired at 298 K.
Data processing was performed with standard Bruker software
XwinNMR and the spectral data are consistent with the assigned
structures. HRMS experiments were performed in the positive
ion mode using electron spray ionization (ESI) source; the LCꢀMS
machine consist of an HPLC Agilent 1290 Infinity System equipped
with a MS detector Agilent 6540UHD Accurate Mass Q-TOF. Ele-
mental analyses were performed on a Fisons elemental analyzer,
model EA1108CHN, and data for C, H, and N are within 0.4% of
the theoretical values.
CDCl3) d 0.90 (d, J = 6.4 Hz, 6H, isopentyl CH3), 1.20 (t, J = 7.1 Hz,
3H, OCH2CH3), 1.40–1.80 (m, 3H, isopentyl CH and isopentyl
CH2), 3.90–4.25 (m, 6H, isopentyl CH2, CH2 andOCH2CH3), 6.90
(dd, J = 2.3 and 8.9 Hz, 1H, H-6), 7.35 (d, J = 2.3 Hz, 1H, H-4), 7.55
(d, J = 8.9 Hz, 1H, H-7). HRMS (ESI) calcd for C16H22NO3S (M+H)+
308.13204, found 303.13267.
4.1.2.2. Ethyl (5-methoxy-1,3-benzothiazol-2-yl)acetate (6c). The
title compound was obtained in manner similar to that used to syn-
thesize compound 6a, starting from commercially available 5-meth-
oxy-2-methyl-1,3-benzothiazole: brown oil, 93% two-step overall
yield; 1H NMR (200 MHz, DMSO-d6) d 1.10 (t, J = 7.1 Hz, 3H,
OCH2CH3), 3.70 (s, 3H, OCH3), 4.05 (q, J = 7.1 Hz, 2H, OCH2CH3),
4.20 (s, 2H, CH2), 6.95 (dd, J = 2.5 and 8.8 Hz, 1H, H-6), 7.40 (d,
J = 2.5 Hz, 1H, H-4), 7.80 (d, J = 8.8 Hz, 1H, H-7). HRMS (ESI) calcd
for C12H13NO3S (M+H)+ 252.06944, found 252.07005.
4.1.1. 5-(Cyclohexyloxy)-2-methyl-1,3-benzothiazole (4a)
Diethyl azodicarboxylate (DEAD) (5.4 mL, 27.24 mmol) was
added dropwise, over 30 min, to a solution of 2-methyl-1,3-ben-
zothiazol-5-ol (3.00 g, 18.16 mmol), cyclohexanol (2.8 mL,
27.24 mmol), PPh3 (7.15 g, 27.24 mmol) and Et3N (2.65 mL,
19.07 mmol) in dry THF (60 mL), placed in a ultrasound bath. The
mixture was sonicated for 72 h and then evaporated to dryness.
The residue was purified by column chromatography, eluting with
CH2Cl2, to give 4a (3.00 g, 67%) as brown solid; 1H NMR (200 MHz,
DMSO-d6) d 1.20–1.50 (m, 6H, cycloxexyl CH2), 1.60–1.70 and
1.85–1.95 (each m, 2H, cyclohexyl CH2), 2.70 (s, 3H, CH3), 4.25–
4.35 (m, 1H, cyclohexyl CH), 6.95 (dd, J = 2.3 and 8.8 Hz, 1H, H-
6), 7.40 (d, J = 2.3 Hz, 1H, H-4), 7.80 (d, J = 8.8 Hz, 1H, H-7). HRMS
(ESI) calcd for C14H18NOS (M+H)+ 248.11091, found 248.10865.
4.1.2.3. Ethyl (5-fluoro-1,3-benzothiazol-2-yl)acetate (6d). The
title compound was obtained in manner similar to that used to
synthesize compound 6a, starting from commercially available 5-
fluoro-2-methyl-1,3-benzothiazole: yellowish solid, 74% two-step
overall yield; 1H NMR (200 MHz, DMSO-d6) d 1.10 (t, J = 7.1 Hz,
3H, OCH2CH3), 4.00 (q, J = 7.1 Hz, 2H, OCH2CH3), 4.25 (s, 2H, CH2),
7.20 (dt, J = 2.5 and 8.9 Hz, 1H, H-6), 7.70 (dd, J = 2.5 and 10.0 Hz,
1H, H-4), 8.00 (dd, J = 5.4 and 8.9 Hz, 1H, H-7). HRMS (ESI) calcd
for C11H10FNO2S (M+H)+ 240.04946, found 240.04980.
4.1.3. Ethyl 2-[5-(cyclohexyloxy)-1,3-benzothiazol-2-yl]-3-(dim-
ethylamino)acrylate (7a)
4.1.1.1. 5-(Isopentyloxy)-2-methyl-1,3-benzothiazole (4b). The
title compound was obtained in manner similar to that used to
synthesize compound 4a, replacing cyclohexanol with isopenata-
nol: brown oil, 89% yield; 1H NMR (200 MHz, DMSO-d6) d 0.80
(d, J = 6.4 Hz, 6H, isopentyloxy CH3), 1.50 (q, J = 6.6 Hz, 2H, isopen-
tyloxy CH2), 1.60–1.70 (m, 1H, isopentyloxy CH), 2.75 (s, 3H, CH3),
3.90 (t, J = 6.6 Hz, 2H, isopentyloxy CH2), 6.90 (dd, J = 2.3 and
8.8 Hz, 1H, H-6), 7.30 (d, J = 2.3 Hz, 1H, H-4), 7.73 (d, J = 8.8 Hz,
1H, H-7). HRMS (ESI) calcd for C13H18NOS (M+H)+ 236.11091,
found 236.10860.
A mixture of compound 6a (2.24 g, 7.01 mmol) and Vilsmeier
reagent, prepared in situ from POCl3 (1.30 mL, 14.02 mmol) and
DMF (1.40 mL, 21.03 mmol), was heated at 90 °C under stirring
for 2 h. After cooling, the mixture was treated with ice-water, basi-
fied to pH ꢁ8 with solid K2CO3 and extracted several times with
CHCl3. The combined organic layers were dried and evaporated
to dryness giving a residue which was purified by column chroma-
tography, eluting with cyclohexan/EtOAc (6:4), to give 7a (2.17 g,
83%) as yellowish solid; 1H NMR (200 MHz, CDCl3) d 1.50–1.70
(m, 9H, cyclohexyl CH2 and OCH2CH3), 1.80–1.95 (m, 4H, cyclo-
hexyl CH2), 3.05 and 3.35 (each s, 3H, NCH3), 4.10–4.25 (m, 3H,
cyclohexyl CH and OCH2CH3), 6.90 (dd, J = 2.3 and 8.8 Hz, 1H, H-
6), 7.50 (d, J = 2.3 Hz, 1H, H-4), 7.65 (d, J = 8.8 Hz, 1H, H-7), 9.05
(s, 1H, CH). HRMS (ESI) calcd for C20H27N2O3S (M+H)+ 375.17424,
found 375.17198.
4.1.2. Ethyl [5-(cyclohexyloxy)-1,3-benzothiazol-2-yl]acetate (6a)
A solution of compound 4a (3.73 g, 15.10 mmol) in a mixture of
1:1 v/v aqueous 50% NaOH and ethylene glycol (45 mL) was re-
fluxed under N2 flux until the starting material disappeared. The
mixture was then poured into ice-water, acidified to pH ꢁ3 with
2 N HCl and extracted several times with CH2Cl2. The combined or-
ganic layers were dried and evaporated to dryness to give crude
aminothiophenol derivative 5a as yellow oil (2.1 g) which was
immediately reacted with ethyl cyanoacetate (2 mL) at 120 °C un-
der stirring and N2 flux. After 15 h, the mixture was cooled and
then treated with Et2O (30 mL). After washing with diluted HCl
and then with diluted NaOH, the organic solution was dried and
evaporated to dryness. The residue was purified by column chro-
matography, eluting with cyclohexane/EtOAc (9:1), to give 6a
(1.05 g, 22% two-step overall yield) as yellowish solid; 1H NMR
4.1.3.1. Ethyl 3-(dimethylamino)-2-[5-(3-isopentyloxy)-1,3-ben-
zothiazol-2-yl]acrylate (7b). The title compound was obtained in
manner similar to that used to synthesize compound 7a, starting
from derivative 6b: yellowish gum, 50% yield; 1H NMR (200 MHz,
CDCl3) d 0.98 (d, J = 6.5 Hz, 6H, isopentyl CH3), 1.25 (t, J = 7.0 Hz,
3H, OCH2CH3), 1.40–1.80 (m, 3H, isopentyl CH and isopentyl
CH2), 3.25 (bs, 6H, NCH3), 4.11 (t, J = 7.0 Hz, 2H, isopentyl CH2),
4.30 (q, J = 7.0 Hz, 2H, OCH2CH3), 6.98 (dd, J = 2.3 and 8.8 Hz, 1H,
H-6), 7.55 (d, J = 2.5 Hz, 1H, H-4), 7.65 (d, J = 8.8 Hz, 1H, H-7),
8.40 (s, 1H, CH). HRMS (ESI) calcd for C19H27N2O3S (M+H)+
362.17424, found 362.17194.
(200 MHz, CDCl3)
d
1.20–1.60 (m, 9H, cyclohexyl CH2 and
OCH2CH3), 1.60–1.70 and 1.90–2.00 (each m, 2H, cyclohexyl CH2),
4.15–4.40 (m, 5H, cyclohexyl CH, CH2 and OCH2CH3), 7.05 (dd,
J = 2.5 and 8.7 Hz, 1H, H-6), 7.50 (d, J = 2.5 Hz, 1H, H-4), 7.65 (d,
J = 8.7 Hz, 1H, H-7). HRMS (ESI) calcd for C17H22NO3S (M+H)+
320.13204, found 320.13287.
4.1.3.2. Ethyl 3-(dimethylamino)-2-(5-methoxy-1,3-benzothiazol-
2-yl)acrylate (7c). The title compound was obtained in manner similar
to that used to synthesize compound 7a, starting from derivative 6c:
yellowish solid, 88% yield; 1H NMR (200 MHz, DMSO-d6) d 1.10 (t,
J = 7.1 Hz, 3H, OCH2CH3), 3.05 (bs, 6H, NCH3), 3.70 (s, 3H, OCH3), 4.10
(q, J = 7.1 Hz, 2H, OCH2CH3), 6.85 (dd, J = 2.2 and 8.8 Hz, 1H, H-6),
7.20 (d, J = 2.2 Hz, 1H, H-4), 7.70 (d, J = 8.8 Hz, 1H, H-7), 8.10 (s, 1H,
CH). HRMS (ESI) calcd for C15H19N2O3S (M+H)+ 307.11164, found
307.10931.
4.1.2.1. Ethyl [5-(3-isopentyloxy)-1,3-benzothiazol-2-yl]acetate
(6b). The title compound was obtained in manner similar to that
used to synthesize compound 6a, starting from benzothiazole 4b:
pale yellow oil, 14% two-step overall yield; 1H NMR (200 MHz,

G. Manfroni et al. / Bioorg. Med. Chem. 20 (2012) 866–876
873
4.1.3.3. Ethyl 2-(5-fluoro-1,3-benzothiazol-2-yl)-3-(dimethyla-
mino)acrylate (7d). The title compound was obtained in manner
similar to that used to synthesize compound 7a, starting from
derivative 6d: yellowish solid, 78% yield; 1H NMR (200 MHz, CDCl3)
d 1.10 (t, J = 7.1 Hz, 3H, OCH2CH3), 2.95 (bs, 6H, NCH3), 4.05 (q,
J = 7.1 Hz, 2H, OCH2CH3), 6.95–7.05 (dt, J = 2.5 and 8.8 Hz, 1H, H-6),
7.50 (dd, J = 2.5 and 10.2 Hz, 1H, H-4), 7.80 (dd, J = 5.5 and 8.8 Hz,
1H, H-7), 7.95 (s, 1H, CH). HRMS (ESI) calcd for C14H16FN2O2S
(M+H)+ 295.09166, found 295.09194.
from compound 7c: white solid, 84% yield; 1H NMR (200 MHz,
DMSO-d6) d 1.22 (t, J = 7.1 Hz, 3H, OCH2CH3), 3.75 (s, 3H, OCH3),
4.25 (q, J = 7.1 Hz, 2H, OCH2CH3), 7.10 (dd, J = 2.5 and 8.8 Hz, 1H,
H-7), 7.35–7.40 (m, 3H, Ar-H), 7.55–7.60 (m, 2H, Ar-H), 7.90 (d,
J = 8.8 Hz, 1H, H-6), 7.95 (s, 1H, H-3), 8.70 (d, J = 2.5 Hz, 1H, H-9).
HRMS (ESI) calcd for
380.09572.
C
21H18NO4S (M+H)+ 380.09566, found
4.1.4.5. Ethyl 8-fluoro-1-oxo-2-phenyl-1H-pyrido[2,1-b]benzo-
thiazole-4-carboxylate (8f). The title compound was prepared
according to the procedure used for compound 8c starting from
compound 7d: yellow solid, 68% yield; 1H NMR (200 MHz,
DMSO-d6) d 1.25 (t, J = 6.9 Hz, 3H, OCH2CH3), 4.25 (q, J = 6.9 Hz,
2H, OCH2CH3), 7.20–7.45 (m, 4H, H-7 and Ar-H), 7.55–7.65 (m,
2H, Ar-H), 7.95–8.15 (m, 2H, H-3 and H-6), 8.80–8.90 (m, 1H, H-
9). HRMS (ESI) calcd for C20H15FNO3S (M+H)+ 368.07567, found
368.07792.
4.1.4. Ethyl 8-(cyclohexyloxy)-1-oxo-2-phenyl-1H-pyrido
[2,1-b]benzothiazole-4-carboxylate (8c)
A mixture of acrylate 7a (0.10 g, 0.27 mmol) and phenylacetic
anhydride (0.13 g, 0.53 mmol), was heated at 100 °C, under stir-
ring, for 5 h. After cooling, the residue was triturated with Et2O
to give compound 8c (0.10 g, 83%) as yellow solid; 1H NMR
(200 MHz, DMSO-d6) d 1.25–1.60 (m, 9H, cyclohexyl CH2 and
OCH2CH3), 1.65–1.85 and 1.90–2.00 (each m, 2H, cyclohexyl
CH2), 4.20–4.50 (m, 3H, cyclohexyl CH and OCH2CH3), 7.20 (dd,
J = 2.4 and 8.7 Hz, 1H, H-7), 7.30–7.50 (m, 3H, Ar-H), 7.70 (d,
J = 8.7 Hz, 1H, H-6), 7.60–7.75 (m, 2H, Ar-H), 8.00 (s, 1H, H-3),
8.75 (d, J = 2.4 Hz, 1H, H-9). HRMS (ESI) calcd for C26H26NO4S
(M+H)+ 448.15826, found 448.16045.
4.1.4.6. Ethyl 2-acetyl-8-(cyclohexyloxy)-1-oxo-1H-pyrido
[2,1-b] [1,3]benzothiazole-4-carboxylate (8g). The title com-
pound was prepared according to the procedure used for com-
pound 8c starting from compound 7a and replacing phenylacetic
anhydride with diketene: yellow solid, 69% yield; 1H NMR
(200 MHz, DMSO-d6) d 1.10–1.45 (m, 9H, cyclohexyl CH2 an-
dOCH2CH3), 1.55–1.70 and 1.75–1.95 (each m, 2H, cyclohexyl
CH2), 2.55 (s, 3H, COCH3), 4.10–4.45 (m, 3H, cyclohexyl CH and
OCH2CH3), 7.20 (dd, J = 2.8 and 8.7 Hz, 1H, H-7), 7.95 (d,
J = 8.7 Hz, 1H, H-6), 8.45 (s, 1H, H-3), 8.70 (d, J = 2.8 Hz, 1H, H-9).
4.1.4.1. Ethyl 2-cyclohexyl-8-(cyclohexyloxy)-1-oxo-1H-pyrido
[2,1-b][1,3]benzothiazole-4-carboxylate (8b). The title compound
was prepared according to the procedure used for compound 8c
starting from acrylate 7a and replacing phenylacetic anhydride with
cyclohexylacetic anhydride. The residue was purified by flash col-
umn chromatography, eluting with cyclohexane/EtOAc (95:5), to
give 8b (40%) as white solid; 1H NMR (200 MHz, DMSO-d6) d 1.10–
1.60 (m, 15 H, cyclohexyl CH2 and OCH2CH3), 1.70–2.00 (m, 8H,
cyclohexyl CH2), 2.75–2.80 (m, 1H, cyclohexyl CH), 4.30–4.45 (m,
3H, cyclohexyl CH and OCH2CH3), 7.20 (dd, J = 2.4 and 8.8 Hz, 1H,
H-7), 7.75 (s, 1H, H-3), 7.90 (d, J = 8.8 Hz, 1H, H-6), 8.80 (d,
J = 2.4 Hz, 1H, H-9). HRMS (ESI) calcd for C26H32NO4S (M+H)+
454.20521, found 454.20294.
HRMS (ESI) calcd for
414.13642.
C
22H23NO5S (M+H)+ 414.13752, found
4.1.4.7. Ethyl 8-(cyclohexyloxy)-1-oxo-1H-pyrido[2,1-b][1,3]ben
zothiazole-4-carboxylate (8h). The title compound was prepared
according to the procedure used for compound 8c starting from
compound 7a and replacing phenylacetic anhydride with acetic
anhydride: orange solid, 53% yield; 1H NMR (200 MHz, DMSO-d6)
d 1.10–1.50 (m, 9H, cyclohexyl CH2 and OCH2CH3), 1.50–1.70 and
1.75–1.95 (each m, 2H, cyclohexyl CH2), 4.15–4.30 (m, 3H, cyclo-
hexyl CH and OCH2CH3), 6.30 (d, J = 9.5 Hz, 1H, H-2), 7.15 (dd,
J = 2.4 and 8.8 Hz, 1H, H-7), 7.85–8.00 (m, 2H, H-3 and H-6), 8.70
(d, J = 2.4 Hz, 1H, H-9). HRMS (ESI) calcd for C20H21NO4S (M+H)+
372.12696, found 372.12468.
4.1.4.2. Ethyl 2-cyclohexyl-1-oxo-1H-pyrido[2,1-b][1,3]benzo-
thiazole-4-carboxylate (8a). The title compound was prepared
according to the procedure used for compound 8c starting from
acrylate 7e and replacing phenylacetic anhydride with cyclohexyl-
acetic anhydride. The residue was purified by flash column chro-
matography, eluting with cyclohexane/EtOAc (95:5), to give 8a
(37%) as yellowish solid; 1H NMR (200 MHz, CDCl3) d 1.15–1.65
(m, 9 H, cyclohexyl CH2 and OCH2CH3), 1.75–2.10 (m, 4H, cyclo-
hexyl CH2), 2.35–2.45 (m, 1H, cyclohexyl CH), 4.30 (q, J = 7.0 Hz,
2H, OCH2CH3), 7.45–7.55 (m, 2H, H-7 and H-8), 7.60–7.80 (m.
1H, H-6), 7.90 (s, 1H, H-3), 9.30–9.40 (m, 1H, H-9). HRMS (ESI)
calcd for C20H22NO3S (M+H)+ 356.13204, found 356.12972.
4.1.5. 8-(Cyclohexyloxy)-1-oxo-2-phenyl-1H-pyrido[2,1-b][1,3]
benzothiazole-4-carboxylic acid (2c)
A mixture of compound 8c (0.29 g, 0.65 mmol) in aqueous 10%
NaOH (7.25 mL) and MeOH (29 mL) was heated at 75 °C for 2 h.
After cooling, the mixture was concentrated, poured in ice/H2O,
acidified to pH ꢁ3 with 2 N HCl and extracted with CH2Cl2. The
combined organic layers were dried and evaporated to dryness giv-
ing a residue which was triturated with Et2O and collected by fil-
tration. Compound 2c (0.21 g, 78%) was obtained as pale yellow
solid: mp 275–278 °C (decompose); 1H NMR (400 MHz, DMSO-
d6) d 1.22–1.55 (m, 6H, cyclohexyl CH2), 1.65–1.75 and 1.90–2.00
(each m, 2H, cyclohexyl CH2), 4.30–4.45 (m, 1H, cyclohexyl CH),
7.20 (dd, J = 2.2 and 8.7 Hz, 1H, H-7), 7.30–7.35 (m, 1H, Ar-H),
7.40–7.50 (m, 2H, Ar-H), 7.67–7.75 (m, 2H, Ar-H), 7.95 (d,
J = 8.7 Hz, 1H, H-6), 8.10 (s, 1H, H-3), 8.80 (d, J = 2.2 Hz, 1H, H-9),
12.70 (bs, 1H, CO2H). 13C NMR (400 MHz, DMSO-d6): d 25.46,
25.51, 31.54, 75.70, 103.48, 107.35, 116.62, 119.96, 122.62,
123.34, 127.86, 128.55, 129.09, 136.01, 136.58, 139.22, 153.63,
156.52, 161.46, 166.36. HRMS (ESI) calcd for C24H22NO4S (M+H)+
420.12696, found 420.12683. Anal. calcd for C24H21NO4S: C,
68.72; H, 5.05; N, 3.34. Found: C, 68.82; H, 5.15; N, 3.14.
4.1.4.3. Ethyl 8-(isopentyloxy)-1-oxo-2-phenyl-1H-pyrido[2,1-b]
[1,3]benzothiazole-4-carboxylate (8d). The title compound was
prepared according to the procedure used for compound 8c start-
ing from compound 7b: white solid, 30% yield; 1H NMR
(200 MHz, DMSO-d6) d 0.90 (d, 6H, J = 6.5 Hz, isopentyl CH3), 1.35
(t, 3H, J = 7.0 Hz, OCH2CH3), 1.60–1.85 (m, 3H, isopentyl CH and
isopentyl CH2), 4.10 (t, 2H, J = 6.5 Hz, isopentyl CH2), 4.35 (q, 2H,
J = 7.0 Hz, OCH2CH3), 7.25 (dd, 1H, J = 2.4 and 8.8 Hz, H-7), 7.30–
7.50 (m, 3H, Ar-H), 7.70–7.75 (m, 2H, Ar-H), 8.00 (d, 1H,
J = 8.8 Hz, H-6), 8.10 (s, 1H, H-3), 8.85 (d, 1H, J = 2.4 Hz, H-9). HRMS
(ESI) calcd for C25H26NO4S (M+H)+ 436.15826, found 436.15590.
4.1.4.4. Ethyl 8-methoxy-1-oxo-2-phenyl-1H-pirido[2,1-b][1,3]
benzothiazole-4-carboxylate (8e). The title compound was pre-
pared according to the procedure used for compound 8c starting

874
G. Manfroni et al. / Bioorg. Med. Chem. 20 (2012) 866–876
4.1.5.1. 2-Cyclohexyl-1-oxo-1H-pyrido[2,1-b][1,3]benzothiazole-
4-carboxylic acid (2a). The title compound was prepared accord-
ing to the procedure used for compound 2c starting from com-
pound 8a. Compound 2a (30%) was obtained as brown solid: mp
266–268 °C; 1H NMR (400 MHz, DMSO-d6) d 1.10–1.50 (m, 6H,
CH2cyclohexyl), 1.70–1.90 (m, 4H, cyclohexyl CH2), 2.79–2.90 (m,
1H, cyclohexyl CH), 7.50–7.60 (m, 2H, H-7 and H-8), 7.95 (s, 1H,
H-3), 8.00–8.10 (m, 1H, H-6), 9.15–9.20 (m, 1H, H-9), 12.70 (bs,
1H, CO2H). 13C NMR (DMSO-d6) d 26.21, 26.85, 32.31, 37.33,
102.87, 119.75, 122.88, 126.82, 127.23, 128.63, 129.93, 132.23,
137.85, 150.57, 162.73, 166.57. HRMS (ESI) calcd for C18H18NO3S
(M+H)+ 328.10074, found 328.10059. Anal. calcd for C18H17NO3S:
C, 66.03; H, 5.23; N, 4.28. Found: C, 66.06; H, 5.26; N, 4.11.
4.1.5.5. 8-(Isopentyloxy)-1-oxo-2-phenyl-1H-pyrido[2,1-b][1,3]
benzothiazole-4-carboxylic acid (2d). The title compound was
prepared according to the procedure used for compound 2c start-
ing from compound 8d. Compound 2d (40%) was obtained as white
solid: mp 268–270 °C; 1H NMR (400 HHz, DMSO-d6) d 0.95 (d,
J = 6.6 Hz, 6H, isopentyl CH3), 1.65 (q, J = 6.6 Hz, 2H, isopentyl
CH2), 1.75–1.85 (m, 1H, isopentyl CH), 4.05 (t, J = 6.6 Hz, 2H, iso-
pentyl CH2), 7.15 (dd, J = 2.3 and 8.8 Hz, 1H, H-7), 7.35–7.40 (m,
1H, Ar-H), 7.40–7.45 (m, 2H, Ar-H), 7.70–7.75 (m, 2H, Ar-H), 7.90
(d, J = 8.8 Hz, 1H, H-6), 8.10 (s, 1H, H-3), 8.80 (d, J = 2.3 Hz, 1H,
H-9), 13.30 (bs, 1H, CO2H). 13C NMR (DMSO-d6) d 22.88, 25.02,
37.76, 66.99, 103.59, 105.70, 115.42, 119.97, 122.55, 123.29,
127.85, 128.54, 129.06, 136.00, 136.54, 139.18, 153.59, 157.91,
161.46, 166.37. HRMS (ESI) calcd for
C
23H22NO4S (M+H)+
408.12696, found 408.12689. Anal. calcd for C23H21NO4S: C,
67.79; H, 5.19; N, 3.44. Found: C, 68.04; H, 5.44; N, 3.04.
4.1.5.2. 2-Cyclohexyl-8-(cyclohexyloxy)-1-oxo-1H-pyrido[2,1-b]
[1,3]benzothiazole-4-carboxylic acid (2b). The title compound
was prepared according to the procedure used for compound 2c
starting from compound 8b. Compound 2b (85%) was obtained as
white solid: mp 307–310 °C (decompose); 1H NMR (400 MHz,
DMSO-d6) d 1.10–1.50 (m, 12H, cyclohexyl CH2), 1.70–1.90 (m,
8H, cyclohexyl CH2), 2.75–2.90 (m, 1H, cyclohexyl CH), 4.30–4.50
(m, 1H, cyclohexyl CH), 7.10 (dd, J = 2.2 and 8.7 Hz, 1H, H-7),
7.75 (d, J = 8.7 Hz, 1H, H-6), 7.85 (s, 1H, H-3), 8.90 (d, J = 2.2 Hz,
1H, H-9). 13C NMR (400 MHz, DMSO-d6) d 23.50, 25.57, 26.36,
27.00, 31.66, 32.76, 37.11, 75.66, 107.48, 111.35, 115.84, 122.27,
122.52, 128.34, 134.11, 139.41, 147.63, 155.67, 162.64, 167.52.
4.1.5.6. 8-Methoxy-1-oxo-2-phenyl-1H-pyrido[2,1-b][1,3]benzo-
thiazole-4-carboxylic acid (2e). The title compound was prepared
according to the procedure used for compound 2c starting from
compound 8e. Compound 2e (65%) was obtained as white solid:
mp 277–278 °C; 1H NMR (400 MHz, DMSO-d6) d 3.80 (s, 3H,
CH3), 7.20 (dd, J = 2.5 and 8.6 Hz, 1H, H-7), 7.15–7.40 (m, 3H, Ar-
H), 7.70–7.75 (m, 2H, Ar-H), 7.90 (d, J = 8.6 Hz, 1H, H-6), 8.10 (s,
1H, H-3), 8.80 (d, J = 2.5 Hz, 1H, H-9), 13.25 (bs, 1H, CO2H). 13C
NMR (DMSO-d6) d 56.03, 104.99, 111.35, 114.93, 120.35, 122.34,
123.22, 127.77, 128.53, 129.06, 136.24, 136.67, 139.21, 153.12,
158.38, 161.48, 166.60. HRMS (ESI) calcd for C19H14NO4S (M+H)+
352.06436, found 352.06413. Anal. calcd for C19H13NO4S: C,
64.95; H, 3.73; N, 3.99. Found: C, 64.67; H, 3.45; N, 4.37.
HRMS (ESI) calcd for
426.17393. Anal. calcd for C24H27NO4S: C, 67.74; H, 6.40; N, 3.29.
Found C, 67.89; H, 6.55; N, 2.99.
C
24H28NO4S (M+H)+ 426.17391, found
4.1.5.3. 2-Acetyl-8-(cyclohexyloxy)-1-oxo-1H-pyrido[2,1-b][1,3]
4.1.5.7. 8-Fluoro-1-oxo-2-phenyl-1H-pyrido[2,1-b][1,3]benzothi
azole-4-carboxylic acid (2f). The title compound was prepared
according to the procedure used for compound 2c starting from
compound 8f. After the purification by column chromatography,
eluting with CHCl3/MeOH (9:1), compound 2f (74%) was obtained
as pale yellow solid: mp 326–328 °C; 1H NMR (DMSO-d6) d 7.20–
7.46 (m, 4H, H-7 and Ar-H), 7.55–7.65 (m, 2H, Ar-H), 7.90–8.10
(m, 2H, H-3 and H-7), 8.87 (dd, J = 2.7 and 11.2 Hz, 1H, H-9),
benzothiazole-4-carboxylic
acid
(2g). LiOHꢂ2H2O
(0.04 g,
0.90 mmol) was added to a solution of compound 8g (0.12 g,
0.30 mmol) in a mixture of dioxane/H2O (4 mL, 3:1). The reaction
mixture was stirred at 40 °C for 4 days, then poured into water,
acidified with AcOH and extracted several times with EtOAc. The
combined organic layers were dried and evaporated to dryness giv-
ing a residue which was triturated with Et2O and filtered. Com-
pound 2g (0.08 g, 88%) was obtained as pale orange solid: mp
299–300 °C (decompose); 1H NMR (400 MHz, DMSO-d6) d 1.10–
1.60 (m, 6H, cyclohexyl CH2), 1.65–1.80 and 1.85–2.10 (m, each
2H, cyclohexyl CH2), 2.70 (s, 3H, CH3), 4.40–4.55 (m, 1H, cyclohexyl
CH), 7.30 (dd, J = 2.2 and 8.8 Hz, 1H, H-7), 8.05 (d, J = 8.8 Hz, 1H, H-
6), 8.60 (s, 1H, H-3), 8.85 (d, J = 2.2 Hz, 1H, H-9), 13.50 (bs, 1H,
CO2H). 13C NMR (DMSO-d6) d 23.41, 25.51, 31.33, 31.50, 75.75,
107.55, 116.94, 118.44, 120.46, 123.75, 139.45, 140.42, 156.92,
159.38, 161.51, 166.11, 195.56. HRMS (ESI) calcd for C20H20NO5S
(M+H)+ 386.10622, found 386.10578. Anal. calcd for C20H19NO5S:
C, 62.32; H, 4.97; N, 3.63. Found: C, 62.12; H, 4.57; N, 4.03.
13.35 (bs, 1H, CO2H). 13C NMR (DMSO-d6)
d 107.30 (d, JC-
F = 30.8 Hz, C-9), 115.00 (d, JC-F = 23.7 Hz, C-7), 122.72, 124.33 (d,
JC-F = 9.5 Hz, C-6), 125.01, 127.925, 128.60, 129.03, 136.43,
136.45, 138.85 (d, JC-F= 14 Hz, C-9a), 153.21, 159.49, 161.37,
164.18 (d, JC–F = 459 Hz, C-8). HRMS (ESI) calcd for C18H11FNO3S
(M+H)+ 340.04437, found 340.04403. Anal. calcd for C18H10FNO3S:
C, 63.71; H, 2.97; N, 4.13. Found. C, 63.93; H, 3.19; N, 3.73.
4.1.6. Sodium salt of 8-(cyclohexyloxy)-1-oxo-2-phenyl-1H-pyri
do[2,1-b][1,3]benzothiazole-4-carboxylic acid 5,5-dioxide (3c)
m-CPBA (0.08 g, 0.48 mmol) was added to a cooled solution of
compound 2c (0.10 g, 0.24 mmol) in mixture of CH2Cl2/acetone
(15 mL, 2:1). The reaction mixture was stirred at room temperature
for 24 h, then a saturated solution of NaHCO3 (10 mL) was added and
the mixture concentrated under reduced pressure. The precipitate
formed was collected by filtration, washed with water and dried.
After treatment of the solid with Et2O, compound 3c (0.02 g, 20%)
was obtained as yellow sodium salt; 1H NMR (400 MHz, DMSO-d6)
d 1.25–1.45 (m, 6H, cyclohexyl CH2), 1.65–1.75 and 1.95–2.00 (each
m, 2H, cyclohexyl CH2), 4.40–4.60 (1H, m, cyclohexyl CH), 7.15 (dd,
J = 2.2 and 8.7 Hz, 1H, H-7), 7.35–7.50 (m, 3H, Ar-H), 7.60–7.70 (m,
2H, Ar-H), 7.87 (s, 1H, H-3), 8.00 (d, J = 8.7 Hz, 1H, H-6), 8.55 (d,
J = 2.2 Hz, 1H, H-9). 13C NMR (DMSO-d6): 23.36, 25.38, 31.32,
75.90, 106.80, 115.26, 119.89, 121.49, 123.96, 128.59, 128.98,
129.19, 135.45, 135.77, 135.93, 138.62, 139.47, 159.73, 162.48,
162.82. HRMS (ESI) calcd for C24H21NNaO6S (M+H)+ 474.09873,
4.1.5.4. 8-(Cyclohexyloxy)-1-oxo-1H-pyrido[2,1-b][1,3]benzothi-
azole-4-carboxylic acid (2h). The title compound was prepared
according to the procedure used for compound 2c starting from
compound 8h. Compound 2h (75%) was obtained as pale yellow so-
lid: mp 297–298 °C; 1H NMR (400 MHz, DMSO-d6) 1.30–1.55 (m,
6H, cyclohexyl CH2), 1.65–1.75 and 1.90–2.00 (each m, 2H, cyclo-
hexyl CH2), 4.30–4.40 (m, 1H, cyclohexyl CH), 6.25 (d, J = 9.4 Hz,
1H, H-2), 7.10 (dd, J = 2.4 and 8.8 Hz, 1H, H-7), 7.85 (d, J = 8.8 Hz,
1H, H-6), 8.05 (d, J = 9.4 Hz, 1H, H-3), 8.80 (d, J = 2.4 Hz, 1H, H-9).
13C NMR (DMSO-d6) d 23.53, 25.51, 31.58, 75.79, 107.23, 111.35,
111.97, 116.32, 119.76, 123.31, 138.27, 138.98, 154.89, 156.53,
162.38, 166.33. HRMS (ESI) calcd for
C
18H18NO4S (M+H)+
344.09566, found 344.09540. Anal. calcd for C18H17NO4S: C, 62.96;
H, 4.99; N, 4.08. Found: C, 62.74; H, 4.71; N, 4.48.

G. Manfroni et al. / Bioorg. Med. Chem. 20 (2012) 866–876
875
found 474.09899. Anal. calcd for C24H20NNaO6S: C, 60.88; H, 4.26; N,
2.96. Found: C, 61.03; H, 4.47; N, 2.60.
of anti-metabolic effect and CulturPlate, Perkin Elmer for
evaluation of anti-viral effect). The microtiter plates were
incubated overnight (37 °C, 5% CO2, 95–99% relative humidity),
yielding a non-confluent cell monolayer.
The evaluation of the anti-metabolic as well as the anti-viral ef-
fect of each compound was performed in parallel. Four-step, 1-to-5
compound dilution series were prepared. Following assay setup,
the microtiter plates were incubated for 72 h (37 °C, 5% CO2,
95–99% relative humidity).
4.1.6.1. 8-Methoxy-1-oxo-2-phenyl-1H-pyrido[2,1-b][1,3]benzo-
thiazole-4-carboxylic acid 5,5-dioxide (3e). The title compound
was prepared according to the procedure used for compound 3c
starting from compound 2e. The free carboxylic acid, obtained after
treatment with 2 N HCl, was collected and washed with Et2O to
give compound 3e (78%) as yellow solid: mp 308–312 °C (decom-
pose); 1H NMR (400 MHz, DMSO-d6) d 3.90 (s, 3H, OCH3) 7.20 (dd,
J = 2.2 and 8.7 Hz, 1H, H-7), 7.35–7.45 (m, 3H, Ar-H), 7.65–7.75 (m,
2H, Ar-H), 7.90 (s, 1H, H-3), 8.05 (d, J = 8.7 Hz, 1H, H-6), 8.60 (d,
J = 2.2 Hz, 1H, H-9), 14.25 (bs, 1H, CO2H).13C NMR (DMSO-d6) d
56.77, 105.77, 111.63, 114.45, 118.25, 124.35, 128.69, 129.27,
129.49, 134.91, 136.10, 136.15, 136.58, 142.52, 159.50, 163.12,
164.91. HRMS (ESI) calcd for C19H14NO6S (M+H)+ 384.05419, found
384.05395. Anal. calcd for C19H13NO6S: C, 59.52; H, 3.42; N, 3.65.
Found: C, 59.85; H, 3.75; N, 3.31.
For the evaluation of anti-metabolic effects, the assay medium
was aspirated, replaced with 75 lL of a 5% MTS (Promega) solution
in phenol red-free medium and incubated for 1.5 h (37 °C, 5% CO2,
95–99% relative humidity). Absorbance was measured at a wave-
length of 498 nm (Safire2, Tecan) and optical densities (OD values)
were converted to percentage of untreated controls.
For the evaluation of anti-viral effects, assay medium was aspi-
rated and the cell monolayers were washed with PBS. The wash
buffer was aspirated, 25
mega) was added after which lysis was allowed to proceed for
5 min at room temperature. Subsequently, 50 L of Luciferase As-
lL of GloLysis Buffer (Cat. No. E2661, Pro-
4.1.6.2. Sodium salt of 8-fluoro-1-oxo-2-phenyl-1H-pyrido[2,1-
b][1,3]benzothiazole-4-carboxylic acid 5,5-dioxide (3f). The title
compound was prepared according to the procedure used for com-
pound 3c starting from compound 2f (47%) as yellow sodium salt;
1H NMR (400 MHz, DMSO-d6) d 7.35–7.45 (m, 3H, Ar-H), 7.60 (dt,
J = 1.9 and 8.5 Hz, 1H, H-7), 7.70–7.75 (m, 2H, Ar-H), 7.95 (s, 1H,
H-3), 8.30 (dd, J = 5.7 and 8.5 Hz, 1H, H-6), 8.85 (dd, J = 1.9 and
10.9 Hz, 1H, H-9). 13C NMR (DMSO-d6) d 108.15 (d, JC–F = 30.2 Hz,
C-9), 11.95, 116.30 (d, JC–F = 23.5 Hz, C-7), 122.87 (d, JC–F = 5.3 Hz,
C-9a), 125.35 (d, JC–F = 10.5 Hz, C-6), 128.74, 128.23, 129.61,
134.74, 135.97, 136.09, 136.74, 141.99, 159.40, 163.04, 165.50 (d,
JC–F = 250.6 Hz, C-8). HRMS (ESI) calcd for C18H10FNNaO5S (M+H)+
394.01615, found 394.01603. Anal. calcd for C18H9FNNaO5S: C,
54.97; H, 2.31; N, 3.56. Found: C, 55.26; H, 2.60; N, 3.53.
l
say System (Cat. No. E1501, Promega) was added and the luciferase
luminescence signal was quantified immediately (1000 ms inte-
gration time/well, Safire2, Tecan). Relative luminescence units
were converted to percentage of untreated controls.
The EC50 and EC90 (values derived from the dose–response
curve) represent the concentrations at which, respectively, 50%
and 90% inhibition of viral replication would be observed. The
CC50 (value derived from the dose-response curve) represent the
concentration at which the metabolic activity of the cells would
be reduced to 50% of the metabolic activity of untreated cells.
A concentration of compound is considered to elicit a genuine
anti-viral effect in the HCV replicon system when, at that particular
concentration, the anti-replicon effect is well above the 70%
threshold at concentrations where no anti-metabolic activity is
observed.
4.2. NS5B enzymatic assay
The cloned gene for HCV 1b NS5B
D
C21 was kindly provided by
4.4. Molecular docking experiments
Dr. Joachim Jaeger, Astbury Centre, RepliZyme Ltd, UK. Recombinant
HCV NS5B was expressed in Escherichia coli and purified by FPLC
chromatography through DEAE and HyperD Heparin columns. As-
says were performed as follows: NS5B (25–50 nM) was incubated
The crystal structure of NS5B RdRp complexed with the inhibi-
tor 9 (PDB ID 2BRK)28 and 10 (PDB ID 3FRZ)30 were retrieved from
the RCSB Protein Data Bank and used as target for our modeling
studies on TSI and II, respectively. Water molecules, except those
involved in mediating hydrogen bonds between the ligands and
the protein, were deleted. Schrödinger Protein Preparation
Wizard33 was then used to obtain a satisfactory starting structure
for docking studies. This facility is designed to ensure chemical
correctness and to optimize a protein structure for further analysis.
In particular, hydrogen atoms were added, and bond orders and
charges were assigned; the orientation of hydroxyl groups on Ser,
Thr and Tyr, the side chains of Asn and Gln residues, and the
protonation state of His residues were optimized. Steric clashes
were relieved by performing a small number of minimization
steps, not intended to minimize the system completely. In our
study, the minimization (OPLSforce field) was stopped when the
RMSD of the non-hydrogen atoms reached 0.30 Å.
in a final volume of 25
8.0, 1 mM DTT, 0.25 mg/mL BSA, 0.25
RNA homopolymeric template (rA)40/(rU)20, 0.25 mM MnCl2 and
M [3H]-UTP (37 Ci/mmol). Samples were incubated 40 min at
l
L in the presence of 50 mM Tris–HCl pH
l
M (30-OH ends) of the RNA/
2
l
25 °C. Twenty microlitre aliquots were spotted on DE-81 filters
(Wathman) which were washed twice in 0.5 M K2HPO4, once in
water and once in acetone. Bound radioactivity was detected by li-
quid scintillation counter (MicrobetaTrilux, Perkin-Elmer). When
inhibitors were tested, they were incubated 5 min on ice together
with the enzyme and the RNA template, before addition of the
[3H]-UTP and subsequent incubation at 25 °C as described above.
4.3. Anti-HCV activity and cytostatic effect
One day before addition of compound, Huh 5.2 cells, containing
the hepatitis C virus genotype 1b I389luc-ubi-neo/NS3-30/5.1 rep-
licon32 and sub-cultured in cell growth medium [DMEM (Cat. No.
41965039) supplemented with 10% FCS, 1% non-essential amino
acids (11140035), 1% penicillin/streptomycin (15140148) and 2%
Geneticin (10131027); Invitrogen] at a ratio of 1:3–1:4 and grown
for 3–4 days in 75cm2 tissue culture flasks (Techno Plastic Prod-
ucts), were harvested and seeded in assay medium (DMEM, 10%
FCS, 1% non-essential amino acids, 1% penicillin/streptomycin) at
Docking studies were performed using the software Glide ver-
sion 5.0.24–27 The prepared structures were used to generate the
receptor grids, with no scaling for van der Waals radii of nonpolar
receptor atoms. The grids were centered on the crystallographic
positions of 9 for TSI and of 10 for TSII.
The structures of our hit compounds (2b, 2c and 3c) and the
known NNIs 1, 9 and 10 were constructed using the Schrödinger
Maestro interface34 and then submitted to Polak–Ribiere conjugate
gradient minimization [0.0005 kJ/(Å mol) convergence]. The car-
boxylic acid moiety of the compounds was modelled as carboxyl-
ate, as predicted by MoKa.29
a density of six 500 cells/well (100
lL/well) in 96-well tissue cul-
ture microtiter plates (Falcon, Beckton Dickinson for evaluation

876
G. Manfroni et al. / Bioorg. Med. Chem. 20 (2012) 866–876
13. Li, H.; Shi, S. T. Future Med. Chem. 2010, 2, 121.
In our study, all compounds were flexibly docked in a stepwise
14. Beaulieu, P. L. Expert Opin. Ther. Pat. 2009, 19, 145.
manner with Glide SP (Standard Precision) and XP (Extra Precision)
as scoring functions. Docking experiments were performed using
0.80 to scale the VdW radii of the nonpolar ligand atoms with a
charge cutoff of 0.15. Two bound conformations per ligand were
retained from SP docking, discarding poses as duplicates when
rms deviation in the ligand-all atoms was less than 1.5 Å and max-
imum atomic displacement was less than 2.3 Å. The obtained dock-
ing poses where then refined, rescored and minimized using
GlideXP. The best XP scoring bound conformation for each ligand
was retained.
15. Hirashima, S.; Suzuki, T.; Ishida, T.; Noji, S.; Yata, S.; Ando, I.; Komatsu, M.;
Ikeda, S.; Hashimoto, H. J. Med. Chem. 2006, 49, 4721.
16. Böhm, H.-J.; Flohr, A.; Stahl, M. Drug Discovery Today 2004, 1, 217.
17. Deore, R. R.; Chern, J. W. Curr. Med. Chem. 2010, 17, 3806.
18. Talele, T. Curr. Bioac. Compd. 2008, 4, 86.
19. Kato, T.; Chiba, T.; Okada, T. Chem. Pharm. Bull. 1979, 27, 1186.
20. Demolition of 2-methyl-5-phenoxy-1,3-benzthiazole exclusively gave the
corresponding disulfide derivative which did not react with ethyl
cyanoacetate.
21. Hodson, S. J.; Bishop, M. J.; Speake, J. D.; Navas, F., 3rd; Garrison, D. T.; Bigham,
E. C.; Saussy, D. L., Jr.; Liacos, J. A.; Irving, P. E.; Gobel, M. J.; Sherman, B. W. J.
Med. Chem. 2002, 45, 2229.
22. Chen, Y.; Bopda-Waffo, A.; Basu, A.; Krishnan, R.; Silberstein, E.; Taylor, D. R.;
Talele, T. T.; Arora, P.; Kaushik-Basu, N. Antiviral. Chem. Chemother. 2009, 20, 19.
23. Beaulieu, P. L.; Bös, M.; Bousquet, Y.; Fazal, G.; Gauthier, J.; Gillard, J.; Goulet, S.;
LaPlante, S.; Poupart, M. A.; Lefebvre, S.; McKercher, G.; Pellerin, C.; Austel, V.;
Kukolj, G. Bioorg. Med. Chem. Lett. 2004, 14, 119.
24. Glide, version 5.6, Schrödinger, LLC, New York, NY, 2010.
25. Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.;
Repasky, M. P.; Knoll, E. H.; Shelley, M.; Perry, J. K.; Shaw, D. E.; Francis, P.;
Shenkin, P. S. J. Med. Chem. 2004, 47, 1739.
Acknowledgments
This work has been supported by Fondazione Cassa di Risparmi-
o di Perugia (ricerca di base 2009, C.P.: 2009.010.00413). We are
grateful to Roberto Bianconi, Stijn Delmotte, Mieke Flament and
Tom Bellon for excellent technical assistance.
26. Halgren, T. A.; Murphy, R. B.; Friesner, R. A.; Beard, H. S.; Frye, L. L.; Pollard, W.
T.; Banks, J. L. J. Med. Chem. 2004, 47, 1750.
27. Friesner, R. A.; Murphy, R. B.; Repasky, M. P.; Frye, L. L.; Greenwood, J. R.;
Halgren, T. A.; Sanschagrin, P. C.; Mainz, D. T. J. Med. Chem. 2006, 49, 6177.
28. Di Marco, S.; Volpari, C.; Tomei, L.; Altamura, S.; Harper, S.; Narjes, F.; Koch, U.;
Rowley, M.; De Francesco, R.; Migliaccio, G.; Carfi, A. J. Biol. Chem. 2005, 280,
29765.
29. Milletti, F.; Storchi, L.; Sforna, G.; Cruciani, G. J. Chem. Inf. Model. 2007, 47, 2172.
30. Li, H.; Tatlock, J.; Linton, A.; Gonzalez, J.; Jewell, T.; Patel, L.; Ludlum, S.;
Drowns, M.; Rahavendran, S. V.; Skor, H.; Hunter, R.; Shi, S. T.; Herlihy, K. J.;
Parge, H.; Hickey, M.; Yu, X.; Chau, F.; Nonomiya, J.; Lewis, C. J. Med. Chem.
2009, 52, 1255.
31. Love, R. A.; Parge, H. E.; Yu, X.; Hickey, M. J.; Diehl, W.; Gao, J.; Wriggers, H.; Ekker,
A.; Wang, L.; Thomson, J. A.; Dragovich, P. S.; Fuhrman, S. A. J. Virol. 2003, 77, 7575.
32. Vrolijk, J. M.; Kaul, A.; Hansen, B. E.; Lohmann, V.; Haagmans, B. L.; Schalm, S.
W.; Bartenschlager, R. J. Virol. Methods 2003, 110, 201.
33. Schrödinger Suite 2010 Protein Preparation Wizard; Epik version 2.1,
Schrödinger, LLC, New York, NY, 2010; Impact version 5.6, Schrödinger, LLC,
New York, NY, 2010; Prime version 2.2, Schrödinger, LLC, New York, NY, 2010.
34. Maestro, version 9.2, Schrödinger, LLC, New York, NY, 2011.
References and notes
1. Shepard, C. W.; Finelli, L.; Alter, M. J. Lancet Infect. Dis. 2005, 5, 558.
2. Bostan, N.; Mahmood, T. Crit. Rev. Microbiol. 2010, 36, 91.
3. Brown, R. S. Nature 2005, 436, 973.
4. Zeuzem, S.; Berg, T.; Moeller, B.; Hinrichsen, H.; Mauss, S.; Wedemeyer, H.;
Sarrazin, C.; Hueppe, D.; Zehnter, E.; Manns, M. P. J. Viral Hepat. 2009, 16, 75.
5. Delang, L.; Coelmont, L.; Neyts, J. Viruses-Basel 2010, 2, 826.
6. Lange, C. M.; Sarrazin, C.; Zeuzem, S. Aliment. Pharmacol. Ther. 2010, 32, 14.
7. Flisiak, R.; Parfieniuk, A. Expert. Opin. Investig. Drugs 2010, 19, 63.
8. Opar, A. Nat. Rev. Drug Disc. 2010, 9, 501.
9. http://www.merck.com/newsroom/news-release-archive/prescription-
medicine-news/2011_0513.html (accessed 09/20/2011).
10. http://investor.shareholder.com/vrtx/releasedetail.cfm?releaseid=573051
(accessed 09/20/2011).
11. Barreca, M. L.; Iraci, N.; Manfroni, G.; Cecchetti, V. Future Med. Chem. 2011, 3,
1027.
12. Watkins, W. J.; Ray, A. S.; Chong, L. S. Curr.Opin. Drug Discov. Devel. 2010, 13,
441.